An emerging literature and preliminary data suggest that circle of Willis anomalies are more prevalent in migraine patients than in the general population. The central hypothesis of this project is that circle of Willis anomalies correlate with alterations in cerebral hemodynamics and contribute to migraine susceptibility and ischemic complications of migraine. The anatomy of the circle of Willis is highly variable and altered cerebral blood flow (CBF) volume has been demonstrated in regions supplied by anomalous circle of Willis vessels. Dysregulation of CBF may allow relative ischemia to develop in the setting of increased metabolic demand related to neuronal hyperexcitability, which may in turn trigger cortical spreading depression, and may predispose individuals with migraine to ischemic lesions and stroke. Using high-resolution 3 Tesla MR angiography, this project will determine the frequency of circle of Willis anomalies in patients with migraine compared to matched controls. Collapsed maximum intensity projection (MIP) images and source images will be interpreted using pre-defined criteria to describe classify circle of Willis anomalies. Patients with migraine with and without aura and matched controls will be enrolled in a 1:1:1 ratio, with 75 patients in each group. The primary outcome measure will be a comparison of the frequency of an incomplete circle of Willis between migraine and control groups. Additional evaluation of individual circle of Willis anomalies, the composite of circle of Willis and intracranial vertebral artery anomalies, and a quantitative measure accounting for the number of anomalies will be performed comparing migraine to controls. Further, this project will correlate circle of Willis anomalies with alterations in regional cerebral blood blow (CBF) based on arterial spin labeled (ASL) perfusion MRI. Regional CBF will be measured both at rest and during a photic stimulation challenge. Cerebral flood flow in vascular regions of interest (ROI) will be quantified, and comparison of quantitative regional CBF and COW anomalies in both migraine patients and controls will be performed. Finally, this project will determine if brain lesions on T2-weighted MRI correlate with circle of Willis anomalies using fluid suppressed T2-weighted MRI. Ischemic changes on MRI will be scored based on location and vascular territory, and white matter lesions will be rated using a semi-quantitative validated scale. Identification of structural alterations in the cerebral vasculature in migraine patients would have several important implications. First, it would provide a developmental mechanism for migraine susceptibility that may lead to further insights into genetic predisposition to migraine. Second, it would expand understanding of potential mechanisms underlying migraine aura and linking migraine with both clinical and subclinical cerebral infarction. Third, it could help to identify the subpopulation of patients at risk of progressive cerebral ischemia so as to target preventative therapies appropriately. It would suggest a role for further diagnostic evaluation to determine migraine mechanism in individual patients, analogous to the current paradigm in ischemic stroke in which determination of stroke mechanism is critical to therapeutic decision-making. Particular pharmacologic interventions may also be more or less appropriate in patients with migraine associated with particular mechanisms, both for prevention and acute treatment. This project will investigate the hypothesis that developmental abnormalities in the cerebral blood vessels contribute to migraine susceptibility and the link between migraine and stroke. If this hypothesis is confirmed, this could lead to new treatments based on a better understanding of migraine mechanisms and might help identify patients with migraine who are at risk of stroke.